The Art of Design for Film: Title, Screens and graphics (2024)

By OLIVER WEBB

Images courtesy of DNEG and AppleTV+.

Created by John Shiban and John Orloff for Apple TV+, Masters of the Air is a World War II miniseries based on Donald L. Miller’s 2007 book Masters of the Air: America’s Bomber Boys Who Fought the Air War Against Nazi Germany. Produced by Tom Hanks, Steven Spielberg and Gary Goetzman – the same team behind Band of Brothers and The Pacific – Masters of the Air follows the 100^th Bomb Group, a Boeing B-17 Flying Fortress heavy bomber unit, as they risk their lives on dangerous missions and aerial battles.

Xavier Bernasconi served as DNEG VFX Supervisor on the show. “I’ve been at DNEG for a long time. They knew I like historical pieces, and Matt Plummer, the executive in charge of the episodic part of DNEG at the time, knew of my love of Band of Brothers,” Bernasconi begins. “I also have experience with larger projects, and this was going to be a very large project for us, with almost 2,000 shots. I’m also fairly technical, and with the virtual production side of things, the company felt that I would be the best option to propose for the client. I had a good conversation with Production VFX Supervisor Stephen Rosenbaum. We got along very well from the start, and it went from there.”

When it came to discussions about the look of the show, an important element was making sure it was grounded in realism. “Trying to be as realistic as possible was the mandate,” Bernasconi says. “Our main source of inspiration was the documentary The Cold Blue, shot by William Wyler during the war. The shooting style was very rigid with not many acrobatic cameras at all. We are not in the co*ckpit of a fighter plane. We are in the co*ckpit of this huge fuselage of metal, and the planes are flying very close to each other for their size.”

Bernasconi and his team wanted to emulate that same style for Masters of the Air. “We tried to replicate that with cameras, as if you were a cinematographer during that time,” he explains. “They would have mounted a camera on the fuselage if it was outside, or they would have had a handheld inside the fuselage when they were inside the plane. They wouldn’t mount a tripod in the middle of an air battle. They would be there with them moving around. That’s what we tried to follow as rules essentially for the camera. For everything else, it was very much trying to be true to the historical data that we had. The prop team created amazing replicas of the B-17 because there are not many left, only three or four. So, they made these amazing replicas that were 3D scanned, and they passed us all these beautiful scans of them that we recreated into our 3D packages.”

Bernasconi looked at additional footage from the War. “The footage was obviously all black and white and very short, but it gave us a good idea of the feeling and of the speed of the planes, and you really get a sense of how little time they had to react. We had a lot of historical data, and from a visual point of view, we used photographs taken during that time. A particularly interesting type of film was used at the time, Kodachrome, and that became a source of inspiration for our look and feel. A lot of it was really archival footage that the production team researched. In the series, it’s shown how at the end of every mission the crew had an ‘interrogation’ where they would go through everything that happened during the flight. Those logs served as the main source of information for all the historical details, and the missions in each episode were modeled on them.”

DNEG was the main VFX partner on the show and was responsible for almost all of the aerial battles. Wētā FX, Whiskey Tree, Crafty Apes and Rodeo FX were some of the other vendors involved. “DNEG is a global company,” Bernasconi observes. “That allows you to pick the best person for the job and not be confined by your studio location. Leveraging our global teams for past projects has always been incredibly beneficial.”

DNEG created nearly 2.000 shots in total for the show. “Everything was modular. We created a lot of rigged parts for bullets, explosions, etc. Then they could all be placed in all planes without having 200 planes with 200 variations; you could mix and match. We shared all of our assets and setups across vendors, which was really interesting. Stephen was the orchestra director of all of it, trying to get everything working. The amount of work he did was incredible. He’s very experienced and was really able to guide us throughout the show and advise on what was important for the storytelling. Sometimes visual effects can take center stage, but for this show we were there to help tell an important story and have the effects be as invisible as possible,” Bernasconi explains.

The sheer amount of work and complexity that went into the show proved to be challenging. “Replicating the look and feel of a flying fortress required some thinking outside the box,” Bernasconi remarks. “The sensation of travel when augmenting co*ckpit shots was a particularly difficult problem to solve. One of the pieces of the puzzle was to simulate laminar flows of air going past the planes. By using various layers of this simulated airflow to perturb the air and the image behind it in the background, you get the sensation that something is happening around the plane but without it being in your face. Of course, we had a fair share of full CG shots. We couldn’t get aerial footage of real B-17s due to the extremely limited number of them that are airworthy, so we used all the references that we had to try to make our CG model as close as possible to the real deal. Some shots, even if we had a B-17 available, would have been too dangerous to try and shoot, so full CG from scratch was necessary.”

DNEG’s Environment Supervisors used satellite images of the entirety of Europe. “The problem was that the images had modern buildings in them,” Bernasconi adds. “So, once we planned out the flight path that was all based on historical data, we painstakingly removed every single highway, every modern building and factory that wasn’t from the time. We repainted in fields, dirt roads and then, for example, when we got to bombing objectives, we would have to perfectly recreate the landscape that was bombed during the raid. Another incredibly complex element of the environments were the cloudscapes. We didn’t want to go for a matte painting approach due to the sheer size of the landscape that we had to cover, so we decided to develop a system that allowed us to create a variety of cloud systems. We had macro clusters of clouds, but each of the clusters consisted of smaller clouds. Realistic clouds are extremely complicated because of the nonlinear nature of their profiles, with at times extremely chaotic patterns. I did a selection of clouds I really liked, and the team created rough 3D shapes based on their profiles. Our effects team then proceeded to simulate the way air moves inside those shapes and freeze it in time. We did that for hundreds of clouds which was a huge amount of work.”

A lot of work also went into the detail of the contrails. “They are physically accurate in the way that they spin around because of the propellers,” Bernasconi details. “Because of the time of day and weather conditions, your position on Earth based on longitude and latitude will give you different colors on your horizon line. We built in various layers of haze. There were so many additional layers of things. We were using contrails to correctly show the flight path and where the planes were coming from. Sometimes, the contrails might not be doing the things you want them to do because of the camera angle, for example. We added a spline that would allow us to bend the contrails how we wanted them to tell the story.”

Getting the minute details precise was an important task due to historical accuracy. Bernasconi notes, “There are lots of people out there who are dissecting the show. I was terrified about being caught cheating or taking shortcuts, but we didn’t. In Episode 1 or 2, they go to Norway to bomb Trondheim. The Germans used a smokescreen, this huge smoke to cover where the city was. We made sure we had the right time of day and that the wind at that moment in Trondheim was going in the right direction. There was actually someone online mentioning that what we did was historically correct and that we got the right direction of the wind. Some people go to that extent of minute details, so it was terrifying to know we’d be scrutinized in that way.”

“Also, for every co*ckpit shot we did the body track of every single pilot, co-pilot and anyone in the nose of the plane,” Bernasconi continues. “We had to remove all of the windows from the virtual production shoot because there would maybe be a camera behind them or a reflection, so we needed to roto out everything and then re-render all those shots for the reflection. That was the level of commitment that we went through. If there was a shot where we only see the pilot, but the camera was positioned in such a way that we see the reflection of the co-pilot, we would also have the co-pilot in the CG plane animated so it would match the shot before and after in the right action and then render it for the reflection.”

“Working on this kind of show was an incredible opportunity, and I was so fortunate to be part of such a hugely talented team. Everyone felt that it was a special show because of the heritage of Band of Brothers and The Pacific. We knew that we were very fortunate to be part of something that’s one of a kind,” Bernasconi concludes.

Audiences were introduced to what appeared to be a ritualistic murder in the Southern heat of the Louisiana delta for the debut of the HBO anthology series True Detective, then headed to the Far North to encounter a bizarre collection of frozen bodies in Alaska for the fourth season titled True Detective: Night Country, which features a cast of Jodie Foster, Kali Reis, Fiona Shaw, Finn Bennett, Isabella star LaBlanc, Christopher Eccleston and John Hawkes. Taking over the roles of showrunner and director was Issa López, who brought on Barney Curnow as Production VFX Supervisor to oversee the visual effects work, with one of the major contributors being Cinesite, which handled 140 shots spanning five of the six episodes that consisted of the series-opening caribou herd, a roaming polar bear and Ennis, Alaska.

“Barney Curnow’s approach to the visual effects on True Detective: Night Country was to have a light touch, respecting the feel and tone of the cinematography and production design,” explains Simon Stanley-Clamp, Visual Effects Supervisor at Cinesite. “He didn’t want anyone to be talking about the visual effects afterwards because that would mean they had been brought out of the world we were trying to create. A lot of the work was world-building because the story takes place in this particular environment of Ennis, Alaska, a fictional town. It’s almost its own character, and we were piecing together this environment using disparate locations and set builds, and using visual effects as the glue to bring it all together, and to expand that world.” Snow was a significant natural element ranging from incidental to a blizzard. Stanley-Clamp explains, “The challenge with this type of work is maintaining continuity throughout the sequence, ensuring that the effect builds over time and conveys the story while also incorporating any nuances in the live-action plates. For instance, a small gust of wind catching the hood of Chief Liz Danvers’ [Jodi Foster] car could initiate a slight change in the simulation. The show is predominantly dark with all the exteriors taking place in darkness, which required some adjustment. We integrated environments and extended Ennis town, often using tiny pinpricks of light to guide the viewer’s attention along a road or to the base of a mountain range. We utilized the torchlight beam from Danvers to illuminate the snowstorm she encounters, with much of the background snow fading into complete darkness.”

Despite the desolate wintry landscape, animals can be found roaming with some of them deciding to visit the streets of Ennis. “Creating the polar bear was a delightful experience,” Stanley-Clamp states. “We meticulously crafted a detailed sculpt, groom and creature effects. Knowing that she would be prominently featured in a long, lingering shot, we initially designed her to be very skinny, as per the script, but Issa later requested that she be bulked up a bit to enhance her presence and create a more familiar silhouette for the viewers. I view the polar bear similarly to how we perceive foxes in the U.K.; they emerge at night, roam our streets and scavenge for food. In the two scenes where the polar bear appears, it simply wanders through, searching for its next meal. The caribou sequence underwent a comprehensive development process, starting with the script, followed by storyboards, animatics and a cut comprised of live-action natural history footage. We combined footage of caribou, deer, reindeer and any other relevant animals to convey the story. Silver Studios created a refined previs in collaboration with Barney and production, which helped solidify the final concept.”

“Subsequently,” Stanley-Clamp continues, “we commenced work on the sequence, developing still concepts for the environment. We began with production footage and expanded upon it, ultimately retaining only the section of terrain where the hunter was positioned. Integrating this element into our final shot granted us complete control over the lighting of the environment, allowing us to have the sun descend below the horizon by the end of the sequence, heralding the permanent night that persists throughout all six episodes. The charging herd of caribou was a combination of hand-animated and crowd-simulated elements, depending on the number of animals in the shot. Super close-up details of the main caribou showcased the groom, complemented by Houdini breath effects and atmospheric elements.”

No major alternations had to be made to the pipeline. “This was the first show to fully utilize USD at Cinesite, ultimately giving greater flexibility to our pipeline, sharing and tracking work through various packages; it was more of a background change to our internal architecture,” Stanley-Clamp explains. “Principal photography took place in Iceland, doubling for Alaska; which meant that by the time we were looking at sequences, there was already a cutting copy and footage for us to look over. After initially bidding on boards and concepts, we pretty quickly received an animatic and previsualization. For the bigger sequences, like the creature animation, Barney tried to get as much in front of Issa for feedback as early as possible, starting with animatics, timing out the storyboards and then previs before anything was filmed. This worked well for the polar bear sequences, which ended up on screen pretty much as the previs. The opening sequence with the caribou, however, went on more of a journey because, unsurprisingly, we couldn’t find any reference of caribou leaping off a cliff to their death! For the look of Ennis and its surrounds, we had a lot of references from [Production Designer] Daniel Taylor’s art department look book, as well as the photography and drone work from Alaska itself. Daniel had a map of the whole town of Ennis, which we used as the basis for the really wide DMPs, and we also took that map and built it out into a wider environment.”

Extensive environment work was required to turn Iceland into Alaska. “Distant mountain ranges were added to the background, leading down to the town, to the outskirts of which extend out to the frozen open expanse of sea and to a perpetual snowy wilderness,” Stanley-Clamp states. “A blizzard approaches around the end of Episode 404, which continues into the following episodes. We matched closely what they had created and filmed on set. They had fake snow with smoke pumped into it, blown in with fans, and we needed to augment and extend that. It looked like fine particulate mist rather than snow and we needed it to match it. We created it using volumetric mist, CG effects, adding in snow particulate to bulk it out, and adjusting the speed, density and depth.”

“Moving away from snow to fire simulations,” Stanley-Clamp remarks, “there’s a sequence towards the end of the season where Danvers is in a warehouse, and there is a fire with sparks floating around. There are embers coming off a bonfire, and when we first see them, they are real. But they need to help tell the story, to lead Danvers through a warehouse, so from a certain point they are visual effects. When she finds an open door, we added the snow blowing through it. Besides weather simulations, digital breath was added throughout the season to reflect the frigid, chilly atmosphere. Bears are warm-blooded creatures, and when we see it peering through Danvers’ driver’s-side window, it breathes on the glass, pausing before walking away. It is an unnerving moment; breath was added to the side window, which evaporates to reveal the bear looking in.”

The biggest challenge was the landscape for opening caribou sequence. “The valley was in five shots that we keep coming back to in the cut,” Stanley-Clamp describes. “We were creating an environment, which was based on a plate that didn’t satisfy the brief. We needed to create something new but not completely different, and there was the uncertainty of not knowing whether the client would like it. We needed to find a sweet spot where it looked similar but better. That was tricky because there were a lot of people who have spent a lot of time finding that location, dressing it and filming it. You need to be sensitive to that but do the right thing for the sequence. Of course, the caribou and polar bear were also difficult. Cinesite has created monsters and otherworldly creatures for countless episodic and feature film productions previously. With fantastical creatures, there is no existing audience expectation about what they might see. However, with realism you have the additional challenge that people think they know what something should look like, which is in many respects harder. I am incredibly proud of the whole team for its work on True Detective: Night Country, which I believe raises the bar in terms of quality and realism.”

By OLIVER WEBB
Images courtesy of Marvel Studios.

Set five years after the events of the 2021 Disney+ mini-series Hawkeye, the origin story of Echo follows Maya Lopez as she returns to her hometown in Oklahoma where she is confronted with her troubled past. To move forward with her life, she must reconnect with her Native American roots and embrace the meaning of family and community. Alaqua Cox reprises her Hawkeye role as Maya Lopez/Echo for the sequel.

Aladino Debert served as Digital Domain Visual Effects Supervisor on the show, which streams on Disney+ and Hulu. “The show essentially had two incarnations,” Debert explains. “We originally started in the summer of 2022 doing concept designs and some R&D for the main sequences assigned to us at the time; things like building assets for the ‘Creation Cave’ and digi-doubles for the main Choctaw characters. Then, suddenly, the show was put on hold by the studio and production was stopped. We feared the show was gone for good.” Debert continues, “But six months later, around April 2023, after I’d finished work on Citadel for Amazon, production restarted with a different client-side VFX supervisor and producer team. Having previously worked on Ms. Marvel, I was familiar to the studio, and given my previous, albeit brief, experience with part one of the show, I started work once again.”

Debert knew that Digital Domain were mostly going to be involved with the main creation sequence for Episode 01. “Those sequences had been shot already, so we had plates, if not necessarily a locked cut.” Debert says. “Production built a relatively small set featuring the Creation Pools and a circle of columns encircling them. The legend revolves around the ancestors, moulded from clay, emerging from the pools and gathering around Chafa, the first to appear. The sequence is rooted in Choctaw lore, where the act of drawing water from the Pools of Life starts a transformation within their world and themselves, eventually turning them into humans. That, in a nutshell, is the essence of the story.”

“In terms of the cave’s look and feel, we had initial concepts from our first designs, but we weren’t sure about our final approach, nor the extent of the virtual set we were going to have to build,” he continues. “My priority, working with CG Supervisor David Liu, was to make sure that our environment felt as real as possible, within reason. Creating this mythical cave was tricky, since we were aiming for realism without good real-world references. This balancing act is always a challenge, so we began experimenting.”

Debert and his team looked at lots of caves for reference. “We also looked at glow worms for inspiration since we wanted a bioluminescent, almost nebula-like appearance to the ceiling. It was an evolutionary process as designs often are. That’s why we studied real-life elements found in caves, such as crystal formations, stalactites, stalagmites and so on. For the exterior shots, the story required the clay ancestors to become human, so we knew that the process of the clay drying up, cracking and revealing the human underneath was crucial. We primarily studied time-lapses of dried mud to understand the process. From there, our effects artists, led by FX Supervisor Jaymie Miguel, went to work.”

When it came to managing the workload with his team, Debert approached the project the same way he always does. “To start a project well-prepared, a great producer is essential, and I had the fortune to collaborate with VFX Producer Ryan Wilk on this show. I’m systematic about certain things, and once you’ve been doing this for a while, you realize most projects have similarities,” he remarks.

“First off, when we receive the turnover of plates, our initial steps involve identifying camera roto/paint needs,” Debert adss. “Camera tracking was straightforward, but, for instance, the women playing the clay ancestors had patches on their nipples, and everyone had underwear on, as one would expect. So, although the makeup team’s work was amazing, there was no way to conceal the fact that you could see what they were wearing. So our roto/paint team, led by Cynthia Trevino, got to work on the task of removing everybody’s underwear. There were maybe 12 or 14 ancestors all adorned with clay makeup and wearing underwear that needed to be removed. We also had to devise a system for color correction that could target specific areas of the clay without appearing too obvious. It was a mix of roto and comp tricks to make sure that we could darken or shift the hue of certain sections. Dealing with a single person who wasn’t moving much would’ve not been a big deal, but once you have 12 or more with motion blur, it became quite the challenge.”

Continues Debert, “And, of course, we had to create the cave. The clients asked for a sense of infinity, wanting it to seemingly stretch on forever. Which was all good, but, of course, we had to build it and then collapse it! That was a predictably long process. Environments Lead Chris Watkins and his team began with what was built on set and then expanded upon it. Ultimately, it took on a cathedral-like look, but with an organic and natural feel, featuring stalactites, stalagmites and a lot of that bioluminescent light. Additionally, the columns, ceilings and floor had veins of lava running through them, so cooling lava was another reference we studied. We wanted to recreate the effect of charred, blackened exterior contrasted with undulating veins of red, glowing bits. The digital set went on forever and ended up being a massive environment that was at times quite tricky to render.”

Debert and his team created around 100 visual effects shots for the show, most of which were for the Choctaw creation sequence. “In the story, Chafa, the first ancestor, drinks from the Pool of Life, initiating the collapse of their underground world, and setting in motion the process that ultimately leads them to become human. As the cave collapses around them, she holds up the ceiling to allow her people to escape. The realism of this sequence was crucial, and I wanted to make sure it felt natural in the way that it was breaking up. The cave couldn’t just collapse into dust. As always, I started by looking at a lot of references of post-earthquake imagery to understand how columns shear and partially collapse without disintegrating entirely. That approach is one that resonates with my work ethos: start with realism as a baseline and then adjust to suit the desired aesthetic or narrative tone. It gives us a solid ground to stand on, no pun intended in this case, while at the same time allows the freedom to deviate artistically when needed.”

Initially Debert’s plan involved simulating the collapse in a linear fashion, starting a distance away from the Pool of Life and ending at its center where the ancestors were. “The idea was to position cameras strategically and allow the dynamics of the scene to dictate the performance,” Debert explains. “But the edit changed throughout the project, and that plan had to be modified since the jumps in the timeline introduced inconsistencies. Thankfully, we had created custom simulations of ceiling collapses or columns crumbling that would be placed tactically to smooth out any continuity issues.”

“A good illustration of the balance between reality and art involved dust. Initially, our shots featured a significant amount of it, limiting visibility into the distance – something that, to me, felt realistic given the context of collapsing rock. However, the client emphasized the need to see more of the cave collapsing, a crucial part of the story, which in retrospect seemed obvious. So, we ended up pulling back a lot of the atmospherics and finding a good balance between realism and creative choices. Compositing Supervisor Eric Beaver and his team were essential in finding that balance.”

Debert and his team also worked on the Biskinik bird, sacred to the Choctaw people and featured on several shots during the creation sequence (the bird was an asset shared with ILM, which was also working on the show). As complex as the bird shots were however, for Debert the most challenging visual effects shot to create involved Chafa’s hands becoming human. “Our R&D for the clay-to-human simulation primarily had focused on larger sections of the body, such as the head and torso, and we were pretty happy with the results,” Debert notes. “The edit had two shots of Chafa observing her hands as the clay cracks and crumbles, revealing her human hands beneath. In the original cut the first shot mostly featured clay hands, followed by an insert of her observing the crumbling, before cutting to the final shot where most of the clay was gone, revealing the actual plate photography of her hands.”

Continues Debert: “We started working on those shots by making sure we had accurate match-moving and determining the balance between CG hands, clay and plate. It was essential for the hands to appear natural, as inside the cave they were only decorated with makeup, resembling normal hands albeit made of clay. So, we aimed for a thin, almost potato-skin-like appearance to avoid a chunky look, which would not be attractive for close-up hands. But then the editor decided to merge those two shots into one that spanned the entire clay-to-human transition, one that would end up being a long close-up shot of hands. Initially, we thought we would simply use a long simulation of the clay crumbling over the combined plate since the original shots were derived from the same plate with an offset. We assumed it would be straightforward and continued working on the simulation side until, three weeks from delivery, we realized we were in trouble: We didn’t have a clean plate of hands for the entirety of this shot, necessitating the creation of full CG hands. Trouble was, we didn’t have accurately modeled CG hands because we hadn’t anticipated needing them!”

“It’s one of those realizations that only hit you in hindsight,” Debert acknowledges. “None of us considered the potential problems until we were dangerously close to delivery. That particular shot turned out to be the most stress-inducing shot in the entire show, catching us completely off guard. It was our infamous ‘pineapple shot.’ Go ahead and Google that! However, the entire team pulled together and came up with very creative solutions to the problems we were facing, and in the end the client was very happy with the result.”

A cultural consultant from the Choctaw people was on set throughout production due to the cultural sensitivities involved. “We had to make sure that the basic designs of the Pool of Life, for instance, or the Birth Mound where the ancestors emerge from after the cave collapses, met with approval from the consultants, so we would present to them for feedback. There were only a couple of times where changes or tweaks were needed, and they were minor, but, nevertheless, it was crucial for us to receive this feedback because it was a sensitive topic, and we wanted to ensure we handled it with care and avoided any missteps.”

Concludes Debert, “When you are working on a show or sequence based on myths or legends, the approach is inherently subjective, and the challenge lies in creating elements that may not exist while ensuring they appear realistic. This requires a fine balance of artistic interpretation and attention to detail. I’ve been fortunate to work on culturally diverse shows lately. In my previous show for Marvel Studios, Ms. Marvel, the lead character, is a Pakistani American teenager with amazing powers. That show was significant to me as it introduced a different family culture, religion and dynamic to the MCU. So, to follow that up with Echo, where the main character, Maya Lopez [played by Alaqua Cox], is a deaf Native American woman with a prosthetic leg and a total badass, was a real joy. It offered a diverse representation within the male-dominated superhero world we’ve grown accustomed to. I learned a lot about Choctaw history from reading and researching for this show, making it rewarding both on a personal level and in terms of the technical challenges, which I always enjoy.”

Brought together by Netflix is the visual effects supervisor duo of Jabbar Raisani and Marion Spates who are collaborating on their third series together, beginning with Lost in Space and followed by Stranger Things and Avatar: The Last Airbender. Their current partnership is an epic adaption of the animated series that was previously tried by M. Night Shymalan in a much-maligned feature film that fans and critics would rather burn or bury.

As for what the live-action adaption could achieve that was not possible for the source material, Spates comes up with one word, “Perspective. The cartoon is 2D and hand-drawn perspectives, while this is all done in 3D with proper depth and scale. You are immersed in the show.” Raisani, who also took on the roles of director and executive producer, is in agreement. “Across the board, we tried to represent if this was real, what would this really look like? Hopefully, that allows for a suspension of disbelief that when you’re watching the animated [series], you know that it’s not real, but we tried to make it feel like a real living, breathing world.”

Spanning eight episodes, the Netflix adaption concludes where the first season of the animated series ended with the Fire Nation laying siege to the home of the Northern Water Tribe as part of its plan to gain global supremacy over the other three nations that have the ability to control water, wind and earth. The main threat is the Avatar, a reincarnated spirit who can master all four elements, who has returned after freak natural accident imprisoned him in a frozen tomb for a century. “We always knew the scale, scope, quantity and quality were going to be difficult because those things never align with the time,” Raisani states. “Specific challenges were Appa and Momo. We knew that we had to nail those characters because how beloved they are, and it took a long time to get them just right. On the Netflix side, working with Ted Biaselli, he was a great resource to look to, talk through the characters and look at the animated series, and he helped us all shape that feeling to get the emotion right for those characters. Once we nailed that, the execution was much faster than we might have anticipated. Bending was hard, too. It was hard to figure out how to do it right. There was a lot of trail and error. It took us awhile to understand exactly what components went into each form of bending.”

Around 3,400 visual effects shots were created for the eight episodes over a period of 18 months by Framestore, Scanline VFX, Important Looking Pirates, Accenture Song VFX, Pixomondo, Image Engine, Rodeo FX, Untold Studios, Outpost VFX, BigHugFX, Cadence Effects, The Resistance VFX, Atomic Pictures, NEXODUS, FABLEfx and DNEG VP. “We looked at the animated series and wanted to mimic everything that we could; however, what do we need to do to ground it in today’s reality?” Spates notes. “We looked at flamethrowers for firebending. For water, we didn’t find a whole lot of reference. We found some water that was in space and a lot of slow-motion buckets of water. You can’t find a water whip online anywhere, but if you do, let us know because we’ll use it for references!” Simulations had to be art directed. “You’re trying to art direct something that is a random event and attempting to use forces to get exactly what you’re after,” Raisani notes. “We understand that you cannot art direct every single drop or drip or element.” A major part of the waterbending recipe was the concept of an underlying force that resembles a twirling barbershop pole. “Jared Higgins, who is a Visual Effects Supervisor, came up with the ‘barbershop pole’ as a way to always move the water,” Raisani explains. “We used it to a degree on the other forms as well, but the water is where you see it the most because what we didn’t want was a ball or orb of water. It always feels like it has a current. That you can see throughout the series.”

Three departments that had to work closely were stunt, visual effects and special effects, all of which had multiple supervisors to manage the workload. “We all start from the moment we are on set,” Spates notes. “We didn’t have an opportunity to come on at the beginning in production, but we were definitely there for the reshoots. The way that Jabbar and I approach things is you get in and work with all of the different departments early on in the process so that way everyone is on the same page of what we’re trying to pull off. They can’t do it without us, and we can’t do it without them.” Easing the communication was an established shorthand. “Fortunately, with people like Jeff Aro, who was one of the stunt coordinators, we worked with him on Lost in Space for years, so there’s an existing relationship that helps to streamline the process,” Raisani remarks. “Nevin Swain was also the prop master on Lost in Space. I know a lot of the crew up in Vancouver from many years of working there.” The special effects team led by Chris Flemington and Mark Gibbard provided some clever solutions. “We had this cool contraption where they could blow wind from a hose,” Spates reveals. “For Aang’s landing, they could just blow the dirt out of the way, so that way we get that effect for free which was great.” Earthbending was mainly digital with exception of some practical debris. “For fire, we had the firelight on their hands to simulate the interactive lighting, and it had to be operated by the board operator, so when someone is bending and doing the firebending all of that has to be timed out and matched exactly,” Spates remarks. “We definitely had some things to figure out for Season 2 because we needed to tighten some areas as far as the interactive lighting and some things that we did on set.”

Classic scenes were recreated, like when Aang is showing off his airbending skills to a group of children and accidentally crashes into a statue. “It was moments like that where you’re trying to emulate the animated series as well as you can,” Raisana observes. “For that particular shot, Gordon Cormier was on a practical driving rig that drove him around, then we do a swap to a full CG version of Aang. Once he comes to camera, he is digital and full CG, crashes and falls to the ground. We definitely used all of the magic of the practical stuff on set as well as full CG stuff, and pulling from this beloved animated series. Moments like that are fun.”

Interestingly, a cinematic franchise that shares the same name but isn’t related provided the location of the Southern Air Temple. “I spent a lot of time in China on another project, but I knew where these amazing mountains [Zhangjiajie National Forest Park] are, which were used in the film Avatar” Spates recalls. “That’s exactly what we wanted because the scale of those mountains is unbelievable. There is a lot of imagery that we can steal from the Internet, which obviously is what helped us to make such awesome Avatar mountains, because that’s what they have been labeled over the course of the multiple shows that have been there. It was awesome.”

Acting alongside the live-action cast were CG characters, with two of the hardest being the air bison Appa and the flying lemur Momo. “There was a huge structure that was covered in fur for Appa that the actors are climbing or riding on top of,” Raisani explains. “There was a lot more of a physical representation of Appa on set and less so with Momo. There is a great scene in Episode 105, which Roseanne Liang directed, where Momo finds a little acorn. This acorn represents the fact that this forest has been burned down is going to be rebuilt. That’s a scene that we worked hard to ensure Momo brought an emotion to his performance and a connection, not only to Katara but to Aang, who is in the position of emotional strife, and make it feel like Momo makes a character choice to give this acorn that he wants to eat to Aang because Aang is struggling and he wants to do something for his friend. It’s moments like that we worked hard to ensure that they are giving a performance as opposed to being cool or cute-looking digital characters.”

A fun creature was the ostrich horse ridden by the Earth Kingdom. “I love the ostrich horses,” Spates remarks. “Accenture Song VFX worked on that creature and did such a good job of bringing the movement of the ostrich into the ostrich horse. It’s unbelievable how [they captured] just little nuance motions of how they move around and walk. Also, there was the challenge of how do we make it into a horse and how do we make fur bend into the tail of a horse? Also, they had all of the armor, too. That becomes a big challenge because all of that stuff has to be simulated because there’s movement in the armor.”

On a different plane of existence is the Spirit World, which can be accessed via the Avatar State. “I won’t get into how the original footage was shot because we weren’t there,” Spates states. “That was one area we could get creative and stylized. Normally, Jabbar and I stay away from stylized stuff because we always try to keep it grounded.” It was a tricky balancing act achieving the proper visual aesthetic. “We were trying to come up with something that felt heightened but also photographic, so we were leaning on a lot of photographic elements like chromatic aberrations, treating it as if it was something that was happening with the lens but was also happening with Aang,” Raisani remarks. “Especially in that first scene where he hadn’t been in the Spirit World, and we were trying to make it feel almost out of focus. It’s overwhelming, and he doesn’t know how to process it. We were trying to get that visually into the footage, but also emotionally connected with what Aang is feeling. In terms of the color, Marion worked with our in-house vendor to figure out how we take this forest in Vancouver and make it feel heightened, but don’t break to where it does not feel like a real place at all.”

A signature effect is the Avatar State when the Avatar achieves maximum power. “As far as the Avatar State and the arrow and eyes, that was something we put in a lot of effort into,” Spates reveals. “We figured out with one of our in-house vendors what that would look like. The tricky thing is you’re putting all of this illumination and light on an actual image of a character. It’s easier when it’s all in CG, but we couldn’t make the CG aspect of Aang be so far different than the physical production footage of Aang. We had to figure out how to illuminate his head. We have a little of subdermal that we put around the arrow and his eyes. How much detail between the iris, pupil and sclera? We spent days and hours [figuring that out].”

Nothing would have been possible without the contributions of the other three Production Visual Effects Supervisors: Jared Higgins, Christopher D. Martin and Alex Gitler. as well as the army of vendors. “Scanline VFX did this incredible sequence of Koizilla wreaking havoc on the Fire Nation, and it’s all done through animation combined with simulations,” Raisana states. “Really complex work that is a combination of character, story, performance and technical complexity. It’s a cool sequence.” Spates agrees with his colleague. “Definitely what I want people to see is Koizilla, which is insane,” Spates says. “You talk about simulation — that is a lot of simulations. It gives me chills to my bones every time I see it. The things that we did to it, also in color, to represent what happens in the animated series has turned out fabulous.”

Raisani has a personal bias. “Another fun sequence is the Aang-Bumi fight, which is part of the block I directed in Episode 104. I leaned heavily on the animated series and tried to do everything I could to represent that animated series in living, breathing form. For every beat that I could, I would grab the animated series and say, ‘We’re going to do this shot and that shot.’ The crew had a fun time shooting it, and we definitely had a great time in post putting that onscreen.”

Mahalia Belo’s remarkable feature directorial debut The End We Start From follows a woman (Jodie Comer) and her newborn child as she embarks on a treacherous journey to find safe refuge after a devastating flood. Based on Megan Hunter’s 2017 novel, The End We Start From is a hauntingly realistic depiction of a dystopian London submerged underwater.

Theodor Groeneboom served as Visual Effects Supervisor on the film. “A friend of mine in London went to film school with Mahalia,” Groeneboom recounts. “Mahalia reached out to me because of him. I used to live in London and worked in a few of the big studios there doing visual effects. Then, I moved back out to Norway a couple of years ago and started my own little company. We do quite a bit of work for the U.K. visual effects companies and some of the independent UK films as well. It felt like an extension of keeping in touch with everything that was happening there. I was involved from the early stages – the screenplay breakdowns, planning on the shoot and throughout the shoot and post-production, so it has been a long journey. I very much enjoyed it.”

The post-apocalyptic elements in the film are a backdrop in the story. “It’s not a visual effects film,” Groeneboom acknowledges. “That’s very much true for how the book portrays it as well. Production Designer Laura Ellis Cricks and Mahalia were very much into making sure the film has some kind of texture to it, not just visually but the way the landscape conveys through the film that it’s not just front and center that everything is happening.”

Explains Groeneboom, “It’s not so much a coherent film that goes from A to B; it sort of drifts in and out of these moments that you take in when you view the film that go away from the whole. We need to treat visual effects like some kind of story-driven element. They are just sprinkled around to provide texture to what is happening to Jodie and her character, which is quite different from making a big scene and point out of it. It just happens to be what they are going through. There are quite a few scenes where we just put stuff in the background that could tell some kind of environmental thing that something has happened and not draw any attention to it. I guess that’s part of the whole textural side of things; they just want to paint the world but not make it completely obvious. I suppose subdued and textual were words that were frequently used about how to make the visual effects integrate. Suzie Lavelle, the DP, was also a big part of that conversation in driving the lighting and the look of everything.”

Groeneboom and his team collected lots of material for visual references and relied heavily on the production design ‘bible’ that Production Designer Laura put together. “We were just looking at real photos of flooded areas such as farmland and cities around Europe and all from recent events,” Groeneboom explains. “It’s all based on real references that are quite current. The concepts from Laura were amazing, and it was sort of a bible – its own locations and references from both the recce and the [actual] places. There’s a bunch of scenes where there are animals trapped in a bit of mud, and you just see those skeletons sticking out. It’s just in the background and you don’t notice it, but it’s everything that can tell some kind of story. Barbed wires being cut for some of the fences, just whatever environmental storytelling they can think of, we tried to put in the background for some of these shots.”

“There were loads of references we were putting into the film,” he continues. “There were references for the textural quality of the rain and how much we should use. Some of it was more on the practical side of things and how to shoot, etc. and elements we want to use. As an overarching production design bible, there was a lot of stuff that came from Laura. A lot of ideas of not strictly stuff they wanted to see in the film, but Mahalia liked the feeling of it. There’s definitely less of the 28 Days Later vibe with the military and trying to keep it a bit more chaotic and grounded in the people around Jodie.”

Belo and Lavelle had a clear vision in their heads of what they wanted to achieve. “It wasn’t specific, but they were after a certain feeling. I think the production design bible picked all the right pieces,” Groeneboom remarks. “It was Suzie who set the real textural film quality to everything with her cinematography. We did really early development, sketches of trying to make things feel like they are underwater, like a suburban submerged in the city. This was all CG stuff that we were playing with beforehand to see whether or not it was doable with a small team and the amount of resources that we had. These were all based off the same references from the production bible.”

The biggest sequence for Groeneboom involved trying to figure out how to do London underwater, specifically Fleet Street. “Underwater in whatever capacity we could do,” he adds. “I think we got there in the end. It was all approached from the same angle as everything else, that it needs to be grounded in reality, and you shouldn’t really pay attention to the effects of it. You should just take in the image as this is something that’s happened. We rebuilt the whole of Fleet Street. I was dangling off a rental bus, one of the old Routemasters, and I was taking photographs of every single façade and building and element and item that I could find. We later modeled them up in 3D for textures and lighting ornament and rebuilt the street from scratch. There are some obvious liberties taken to make sure that lighting looks as good as it can, so there are gaps in-between the buildings just to put nice eye lights on every other building. Once the light hits a certain angle on Fleet Street, it just becomes completely obscured by the buildings.”

Fleet Street proved to be a challenging task as it’s an iconic street in London that people are very familiar with. “There are so many great photo references of it as well,” Groeneboom details. “It was a matter of bouncing between what Mahalia felt was real, or what she would accept being real, and what Suzie thought of the texturing and lighting of the scene and how she would approach it from a practical point of view. Obviously, she can’t light an entire street, but how as a DP would she approach it from a practical point of view? We had lots of discussions trying to figure out the right angle for the sun, but also blocking out certain elements to create interesting patterns on one side and having the other side a bit more muted. The opening shot of that scene was quite challenging as well, just trying to make it feel like London specifically. It was shot on a little greenscreen.”

There were around 120 visual effects shots in total, with 13 or 14 of those being Fleet Street. “The film is quite slow-paced, especially in terms of number of cuts in the film. Every single little detail and item on Fleet Street was modeled up and painstakingly created. There are some hints of hope in this scene as well. For example, there are few people dotted around in the windows. The bus was a challenge as well because the chassis of the bus was not favorable to any particular lighting. If you see the Routemasters buses in the city and take pictures of them, they are just uniformly red. It’s very hard to see any shading on them. Getting some prospective lighting on them proved to be a little bit difficult, so we had to exaggerate the amount of light and shadow that the material actually has. It’s a combination of plastics and metal, so that was a bit of a faffle.”

Continues Groeneboom, “[Lighting was also an issue on] quite a few shots where there is flooding on some of the signs, but it’s all CG. We rebuilt a lot of environments and tried to make it all as integrated as possible. For instance, the subtly of some of the elements, like dead animals floating. Maybe you don’t pick it up when you watch the film for the first time, but there are quite a few of these images where they are driving and there is stuff in the background. I quite like that because a lot of films I’ve worked on previously have been blockbusters and the effects are front and center, but here they are subdued in the background. One of my favorite shots is of a traffic jam. There is a giant boom mic on the windshield of the car, which was a bit of pain to remove. We just extended the whole background with a bit of the M25 in the background, and there are fires and fire trucks and lots of things happening, but you probably don’t see it. In terms of rain enhancements, it was just putting more in. On the day, you can only get rain so far close to certain things before you have to do some augmentation with visual effects because of rain. Working with real effects elements is always a bit unwieldy. We also worked on the little baby bumps as well, which are partially prosthetics and partially visual effects.”

Groeneboom and his team had ample support from every department in pre-production and on set. The riggers and the gaffers were especially helpful in pulling up greenscreens if needed and accommodating for the lighting and tracking markers whenever Groeneboom and his team were able to be on set for supervision. “We were there for the most important days, but they would take our potential work into consideration by just phoning us up and asking if we need tracking markers here, for example,” he adds. “So, in terms of on the shoot, we were quite welcome and an integral part of solving some of these shots. In terms of the post-production side, it was me and my company [Rebel Unit in Bergen, Norway] doing it. SunnyMarch were producing the whole film, and they were our client for the job. We are a small team of 10-11 people. I think we had six people on this at the most. The idea of my company is most of the people working there have worked in larger facilities before, so we are trying to move away from the idea of thinking large pipelines and overcomplicating, or over-engineering things, which we are trying to do as much as we can in off-the-shelf software and just being a bit more nimble about the approach. I’m quite happy with a small team of six people doing the work. There were two of us, including myself, doing the modeling for the Fleet Street elements.”

Groeneboom concludes,“Given the fact we are a small team and what we were able to achieve, I’m quite proud of the work we did. I really enjoyed working with Mahalia. It’s her first film, and I think she’s approached it in such an interesting and inspiring manner. I’m very interested in seeing what she does next. It was also great working with Suzie, as she’s lovely to work with, and working with the production team was one of my favorite parts. Reading good reviews, and the fact that the film has an important backdrop about the state of the world is interesting. From a visual effects point of view, you tend to go with more lackluster ideas of what apocalyptic visions happen to be. I thought this felt more real and important. Also, watching Jodie perform was really cool. She’s absolutely phenomenal.”

As Akira Kurosawa was a major influence on George Lucas’ The Hidden Fortress in particular, Zack Synder was inspired by Seven Samurai for what was initially meant to be a Star Wars pitch that has since been retooled for Netflix as a multiplatform original IP. A feature film was shot that has been divided into two parts with the first being Rebel Moon: Part One – A Child of Fire followed by Rebel Moon: Part Two – The Scargiver. Participating in what Synder has termed a “giant atmospheric space adventure” is Visual Effects Supervisor Marcus Taormina who previously collaborated with the filmmaker known for speed ramping, hyper-real, painterly compositions and lens flares on another Netflix production, Army of the Dead. “What has been nice about working on both movies simultaneously, both shooting them and doing post-production, is that we’re looking at both movies at the same time,” Taormina states. “A lot of what is set up in Part One we use in Part Two.”

Overall, 1,380 visual effects shots were created by Framestore, Luma Pictures, Mammal Studios, Rodeo FX, Scanline VFX and Wētā FX for Rebel Moon: Part One – A Child of Fire, which revolves around an adopted daughter of a despot standing up against him by assembling a gang of notorious renegades to protect a planet that she now calls home. What makes the production somewhat unusual is that Synder doubles as his own cinematographer. “It’s nice to have a director/DP because I only have to go to one side of the set versus splitting time,” Taormina notes. “Days of Heaven was a huge inspiration for this movie, which means organic filming, daylight dependent, lots of lens flares, and we also had a custom one-of-a-kind anamorphic package that Zack created for the film, which in itself was a huge challenge. What was nice was that I could go to him about things that I needed. ‘I love the lens flares, but I need you to do a clean pass as I have to erase that flare and put it back over the work later on.’ He was understanding to that and granted me those opportunities on set.”

Skies set the tone for scenes, and Snyder sent a whole library of them to Taormina. “It would always be, ‘I like this and that reference,’” Taormina explains. “I would pull them together and go, ‘What do you like about this one?’ I would smash them all together. We had a lot of discussions about the gas giant Mara at the beginning of the movie, which is not a sky, but it’s [related] enough that it’s an important part. Zack found a colored look that he liked that was a dirtier orange. Obviously, he had to put the lightbox up there, which was headache in itself because the flare contaminated the lens a lot. There was a lot of compositing needed to put those shots together, but it adds to the believability because when you get this dirty orange wash over Kora [Sofia Boutella] it feels cohesive.” Over 38 worlds had to be conceptualized with the main ones being Veldt, Neu Wodi, Daggus, Castor, Sharaa, Gondival and Motherworld. “It starts with our production designers, Stephen Swain and Sefan Dechant; they sent a lot of reference packages our way. Obviously, if there were practical pieces, we would try to infuse those back into our digital worlds. But trying to make them unique yet familiar was a challenge. Atmospherics, the mood and lighting, all of those things were important and played a role. When going to Sharaan we meet King Levitica. It’s moody because it feels like we’re not supposed to be there as the viewers,” Taormina remarks.

For arid Neu Wodi, the ranch was a real location; however, when the creature known as a Bennu takes flight, the entire environment becomes digital. “We talk about how we wanted to get a lot more claustrophobic,” Taormina states. “When we fly through the spires, it feels like they’re closing in on us, and then the Bennu [nicknamed Beatrice] smashes Tarak [Staz Nair] off onto the cliff. We have to make sure that when Tarak jumps that he jumps down into a huge ravine, which feels threatening. We let it open back up towards the end where there’s this majestic scale and beauty to it. Designing that was fun.” The Bennu harkens back to Pegasus. “It’s neither a raven or crow but both at the same time, and a gryphon, too.” Stunt performer Albert Valladares was placed in the middle of a parallelogram with his colleagues holding on to ropes that were attached to his backpack to simulate what it would be like trying to restrain a rearing Bennu. “Every take I was like, ‘Zack, hang on. I’ve got to give a note.’ I’m giving notes to special effects and creature stunt performers while Zack is giving notes to Staz Nair. On occasion I would say, ‘Staz, just imagine that at this moment that you’re going to get ripped apart, and now you have a subtle moment where you get to interact with him. Live in that moment. And also let Albert do some of the performances and lead you, in a sense,” Taormina says. Nair was subsequently captured sitting on a gimbal setup in a parking lot for the aerial sequence. “I said to Zack, ‘I know that we have these anamorphic lenses, which are great. However, let’s do one camera anamorphic and shoot the other three spherically. It will be super sharp, but don’t worry, we’ll add our optics later on to it to make it feel anamorphic.’ We basically reanimated and recomposed all of those shots and scaled them down so that it feels like he’s flying through.”

Influencing the aesthetic of the cobalt-mining planet Daggus was Blade Runner. “I wanted it to feel dark, dingy and moist,” Taormina states. “A lot of narrative and production design determined that environment as we had a lot of built pieces to the set.” Dwelling in the basem*nt level is a native spider/humanoid hybrid called Harmada (Jena Malone) which abducts a child and in doing so comes into conflict with Nemesis (Bae Donna). “The stuntvis or previs was a combination of digital shots and stunts in motion capture suits. Because that space was so small, stunts had to be careful about the movements. Jena Malone is on this huge swivel rig, which is best described as a cart that she is strapped to because we didn’t want her body moving too much, as we needed to see her just below the navel or bust to make it cohesive. We had three to five stunt performers swiveling the rig while having it going up and down to the performances. They also have noodles that are interacting with Donna. It was a harmonious integration of everything together because we didn’t want to replace Jena as her performance was too great.” Nemesis wields two swords that avoid being replicas of lightsabers. “When I had that initial conversation with Zack. all we knew is that we were going to have acrylic rods with LEDs in them that had a warmer amber color. I had this light painting of streaking sparklers. I was like, ‘That’s cool. I think we can do that almost like a synthetic shutter or delayed shutter on the swords, add a heat signature and smoke and then add what we call ‘sword popcorn’ as well. It’s those sparks that came off of it.”

Anthony Hopkins voices a robot called JC-1435, also referred to as ‘Jimmy,’ that decides to participate in the rebellion against the oppressive Imperium led by Kora. “I asked Zack if we could video record and do a ADR scratch session with Anthony Hopkins,” Taormina reveals. “If you don’t have your performer listening to the way that Anthony delivers the lines then there may be a mismatch, and when you put them all together in the end, your brain is going ‘something is wrong or odd here.’ The inflections and body movements are not right. We made the Jimmy suit for Dustin Ceithamer that had chest plates in the front and back, shoulder pads, a face plate and some hands, but the hands had to be replaced digitally. When we got into post, Dustin’s performance and Zack’s direction were so great and minimal that it was less robotic and more human. I actually mandated that when we could, which is about 90% of the shots, to keep the practical chest and face plates, which is a lot more difficult to do because it was so beautiful in the way Zack captured them in the available light.”

Another interesting approach was for the restraining devices that have a mechanical base resemble a crab. Explains Taormina, “Those are called ecto-shackles in the script, but we named them ‘Beetlejuice Chairs.’ We have stunt performers grabbing the cast when they get thrown back, and then we swap out and put the practical prop in there. Like for Gondival or in Providence when the gentleman gets captured, we do the stunt followed by the digital version of it and then swap it out, and we had special effects create this RC-controlled base of the ecto-shackle. We put him on that with only the spine and added all of the digital pieces of it walking. Again, with your brain you’re trying to do the trickery of ‘what’s real and not.’ That’s a great example of stunts, props and special effects doing a fantastic job. We get the plates and go, ‘Let’s make this look cool.’”

Producer and long-time David Fincher collaborator Peter Mavromates extend their partnership in the The Killer where an assassin seeks revenge after a botched assignment. The Netflix feature consists of 900 digitally-augmented shots that range from shortening the tail of a dog to CG airplanes, tasked to a vendor list that includes Ollin VFX, Artemple-Hollywood, Savage VFX and Wylie Co. as well as an in-house team. “Visual Effects Compositor Christopher Doulgeris and I will go into the color bay with [Colorist] Eric Weidt and talk about some issue that we had,” Mavromates explains. “Even sometimes if it’s an outside vendor, we’ll focus to help problem-solve. It’s this wonderful and fluid atmosphere, and it works for David Fincher because he’s always got ideas flowing. He doesn’t want to be on a clock at a facility where you’ve got from 2 p.m. to 5 p.m. and then it’s overtime. There’s none of that. David will walk the halls and stop in on people to check on stuff.”

An unusual visual effects situation arose when a guard dog pursues The Killer, played by Michael Fassbender, after its owner has been murdered. “The dog gives a vicious performance but had a tail that is probably about 12 or 14 inches long, which drove David crazy because when it wagged, he looked too cute!” Mavromates laughs. “David calls me in and says, ‘We’ve got to get this tail down to two inches.’ This is a night scene, so in terms of the type of work that you have to do on manipulating an image, it was tough footage. We had 37 shots, and for that we worked with Ollin VFX in Mexico, and they doctored the tail. When you look at the movie and see that tail, there is another 10 inches or so that you’re not seeing anymore!” What has become more common is the reframing of shots in the DI. “That is something David discovered while we were still shooting on film for Panic Room,” Mavromates remarks. “Once you had all of that film scanned and you’re in the DI suite, there is an opportunity to improve on the headroom. You couldn’t move it right or left that much because of the way it was shot on the negative, but you have a lot of north and south. We did about 100 shots then, and that number has continued to go up where it’s more than 50% of the shots in recent movies where the framing is adjusted.”

Considering that the entire first reel of the movie has the protagonist surveying the building across the street in Paris, one would have thought that Rear Window would have been an influence. “It’s astonishing how little we talked about that movie,” Mavromates notes. “The movie that David referenced and has nothing to do with the look of it is Le Samourai in terms of the tone and what the character is.” The daytime building in Paris actually exists and has been featured in Emily in Paris. “The initial shooting was in Paris and was in that square,” Mayromates explains. “We had eight cameras rolling so that David could capture the images of the people walking in the square and the façade of the building in the daytime. You could capture them simultaneously with different lens lengths so that the action matches perfectly because it’s literally the same take. Later, in New Orleans, they shot the individual window settings all laid out on a stage for the night scenes. Based on the daytime footage of that building and some nighttime plates, Artemple built an element where they put those windows in that were shot in New Orleans, then added tinted glass on the foreground.”

A digital double was created for the scooter escape of The Killer through the streets of Paris after the botched assignment. “The scooter and Michael Fassbender are all CG,” Mavromates reveals. “They did shoot Michael Fassbender on a scooter. That was in our early edits. Then Wylie Co. came in and slowly replaced everything. The background is a photographic plate behind the digital character and scooter in the foreground. If you were doing that all from scratch, that’s a big ask. At least when you have those plates, a lot of lighting decisions are made inherently.”

Even the Fitbit that Fassbender wears got a facelift. “There might be a few Fitbits that are actually photographic,” Mavromates observes. “I doubt it, because we created an interface in post for it. The interface that you see is not one that is exactly right for a commercial product as with the little music player that he has. Somebody asked me, ‘What is that MP3 player? Is that the Microsoft one? What is that?’ I answered, ‘No, that is the Fincher pod.’” The Molotov co*cktail that Fassbender throws was also digitally augmented. “Special effects put a wick in the bottle that had these LED lights, which were golden, and Ollin VFX went in and put the flame over that. What was astonishing is Fassbender threw the bottle that far and it landed a little bit to the left of the door. I couldn’t do that, for sure!”

Not since Neill Blomkamp released District 9 in 2009 has an international production received receive Academy Award and VES Award nominations for its visual effects work. But Takashi Yamazaki has repeated the feat by stomping through the box office beyond Japan along with an iconic kaiju that has been cinematic staple since 1954. Godzilla Minus One revolves around a World War II kamikaze pilot suffering from survivor’s guilt having a re-encounter with the title character, which has gone through further mutation because of American nuclear tests at Bikini Atoll.

Visual and special effects have dramatically evolved like the creatures in the Godzilla franchise. “In the current digital era, we tried to use technology that could only be used digitally,” notes Yamazaki, who was previously responsible for the live-action adaption of Parasyte. “We have a lot of close-up shots of Godzilla to instill fear in the audience because it’s quite rare in Godzilla films for Godzilla to appear in the same scenes as people. We made it possible because of the high level of detail we included in the CG. On the story side, 1954 Godzilla does a great job balancing the human drama with the Godzilla scenes; therefore, we were mindful in trying to have strong story and character development, and to make sure that it’s woven together with what Godzilla is doing on screen.”

Along with being the director, Yamazaki was the Visual Effects Supervisor on the project. “Let me add screenwriter to that as well! Being all three had its benefits, although having said that, when I went to shoot on location, I wanted to ask the writer why he wrote in that specific scene because it was so difficult to shoot! When we went into post, I wanted to ask the director why did he shot the scene in that particular way because it made the visual effects that much more challenging! But of course, I only have myself to blame for all of it!” The different roles had an influence on each other. “Normally, when I write a screenplay, I have to pay some consideration to the visual effects team. Can they achieve this scene? However, in this instance I decided to trust my future self for the sake of efficiency. In post-production, because the director was the one and same as the visual effects supervisor, it allowed for more trial and error in the same amount of time. We were able to avoid any miscommunication or difference in creative direction when it came time for approvals.”

“We watched various Godzilla movies again and learned what makes everyone think, ‘This is Godzilla,’” Yamazaki explains. “We analyzed a great deal of photos and videos for historical background. The assistant director’s team collected a variety of background materials, and I remember being terrified that we could no longer use existing materials. Once we had a clear picture of what was going on at the time, we couldn’t just fudge it.” Storyboards drove the design process. “I drew everything that involved visual effects. It was a huge amount of work. The assistant director [Kôhei Adachi] and Kiyoko Shibuya [Visual Effects Supervisor at Shirogumi, which was the sole vendor] were very impatient with me! Basically, there was no such thing as concept art. Based on the storyboards, I set up the scenes and created previsualization with some staff members using simple CGI. From that point on, I sat next to the CG artists and gave them direct instructions if they were not going in the direction I was aiming for. If something is closely related to the shooting, I made it before the shooting starts. However, if it was necessary to include it in the editing, after the shooting is over [for example, a full CG shot], I made postvis in the same way to set the rhythm of the editing.”

The overall silhouette of the creature is based on the previous Godzilla suits. “The dorsal fin is more acute and the legs are thicker to give a more ferocious impression,” Yamazaki remarks. “We tried to heighten the fear by making Godzilla closer to the characters, so we included many fine details to allow the camera to get closer to the characters.” No motion capture was utilized in the animation of Godzilla. “However, to help define Godzilla’s look, the animator and I spent a lot of time testing Godzilla’s walk. In Shin Godzilla, Godzilla’s posture feels very straight and uptight. In Hollywood, Legendary’s Godzilla feels more aggressive, like an animal ready to pounce. But we wanted something different from both of those. In Japan, Godzilla represents both God and Monster, so we wanted its movement to feel almost divine or God-like. We adjusted the height of its waist, how it moves and its posture many times before arriving at its current design,” Yamazaki states.

As for the world-building, it was important to transport audiences back to 1947. “We collected a large number of photos and videos from that time and were conscious of recreating the atmosphere of that period,” Yamazaki explains. “We could not use any existing open sets, and we did not have the budget to construct new buildings, so we created a composite of a digital building on a simple road set. It was difficult to make the two fit together.” The visual effects shot count is misleading. “Although there were 610 cuts, screentime amounted to two thirds of the entire film. Production time was roughly eight months after the shoot was over and in full swing. Although several people were involved in modeling and scene design before the shoot.”

The first ocean battle was the most complex scene to execute. “We wanted to shoot this on location at sea because I felt it gave the picture this cool documentary style,” Yamazaki reveals. “What I didn’t expect is that everyone got seasick, and the weather was quite unstable, which made filming difficult. Once we took the footage back to the office, the natural waves that we captured were both beautiful and complex, which made it difficult making a giant creature swim through it creating its own waves. With that said, filming on one location and overcoming this challenge unified the team, and that natural imagery we were able to capture made the shot more powerful and convincing. Would I write an ocean scene into my next screenplay? That’s debatable!”

Simulating the ocean was hard because of the amount of data and work to make it appear believable. “A young staff member showed us a simulation of the ocean that he had made as a hobby using his home-made computer,” Yamazaki recalls. “It was so good that we rewrote part of the scenario and increased the number of ocean scenes considerably. However, in the latter half of the work, I regretted why I did that. We didn’t have a server with plenty of data to store it, so we had to make do by deleting cuts as they became available. I was astonished when I was told that the total amount of data exceeded one peta!” A personal favorite is Godzilla emitting a heat ray and destroying the entire Ginza area. Yamazaki comments, “Including the gimmick of the dorsal fin and the depiction of the area after it is destroyed, I believe I was able to create a heat ray that is more powerful and more horrific than ever before; that is a proper metaphor for the atomic bombing.”

The AppleTV+ series Monarch: Legacy of Monsters provides the backstory for the mysterious organization formed upon the discovery of kaiju through a family drama where two half-siblings learn of each other’s existence and their father’s connection to Monarch. The MonsterVerse production created by Chris Black and Matt Fraction on behalf of Legendary Entertainment, Toho Company and Warner Bros. Entertainment consists of 10 episodes that required Visual Effects Supervisor Sean Konrad to craft over 3,000 shots with the help of Rising Sun Pictures, Rodeo FX, Framestore, FuseFX, Outpost VFX, Crafty Apes, Wētā FX, MPC, Storm Studios, Vitality VFX, BOT VFX, Mr. Wolf, Scarab Digital, The Third Floor, Proof, MPC Visualization and an in-house team.

Integrated into the narrative are previous MonsterVerse installments that appeared on the big screen. “What’s interesting about the show is the scene in the beginning of Episode 101 has this Kong: Skull Island [2017] classic adventure movie tone that is a bit off-the-wall crazy and every second has a new amplification of stakes through the action,” Konrad notes. “Then you get back to Kate Randa’s [Anna Sawai] perspective later on in the episode, and you’re going to Godzilla [2014] and you’re having that serious tone of this is a city being destroyed. You need to communicate those ideas visually in a way to people that is meaningful and serious. Then you have the final creature scene in the episode, which is a bunch of new monsters coming out of the ground and attacking our protagonist. Each one of them has a different tone.” Do not expect shots from the kaiju point of view. “A lot of the series is constructed around a subjective point of view of our characters experiencing the action from their perspective and reinforcing that with the visual effects design. A lot of times we knew the kind of creature we wanted and the terrain was immaterial to what the creature was doing, but it did influence how and where we shot it,” Konrad says.

Taking advantage of the physical location rather than rely heavily on bluescreen was the mandate. “We went to a bunch of locations in Hawaii for Episode, 101 and one of them was Lānaʻi Lookout [on O’ahu],” Konrad recalls. “It is this beautiful volcanic rock cone that points out into the ocean. If you’re running away from a monster that’s a great place to be heading towards. We looked at the terrain and the action we wanted to plan. There is a big bamboo forest in the corner of this landscape, and we let that take you through Bill Randa [John Goodman] being chased by the Mother Longlegs spider from Kong: Skull Island to a single point on that landmass, and this giant crab, which is made from the same volcanic rock, comes out of the ground. It was difficult because we prevised the scene based on some storyboards before scouting it, which is always a dangerous thing. We didn’t have a crab designed, so we grabbed The Third Floor graphic crab out of their archive and animated that for the previs. Simultaneously, I was doing the concept for the crab itself. All of that happened within a five-week period where we concepted, re-prevised and re-storyboarded the whole thing. It’s so much better than trying to shoot that into a bluescreen. It’s a hard process and definitely time-consuming. People did some brilliant work to get that done. We wanted the characters to feel in the action.”

Television production happens at a quicker pace than movies. “When you’ve got one movie, you have over six months to do an hour and a half episode as supposed to 30 days to shoot a movie,” sates Special Effects Supervisor Paul Benjamin. “The setups have to be doable to make it into TV land. We had a bit of prep time for the first two episodes. It helps if the directors are on beforehand to know if you have some bigger builds. However, it’s always hard to get the directors before the episode starts.” Some visual research was done regarding the MonsterVerse. “I took a quick view of what they’ve been up to and been doing. When you watch the movies, sometimes it’s hard to figure out how they did everything, or how you block it out and film it. You get a general sense of what you’re up against or what they’re going to be looking for. But as far as pulling builds from watching the movies, it’s quite difficult,” Benjamin adds. Atmospherics were not a significant part of the visual language. “We did a couple of episodes of heavy smoke when they went to the Lost Lands to give a different look to the environment. We definitely did some snow but didn’t do a lot of exterior snow dressing, except for the base camp.” The weather was not always agreeable. “We did a snow dress and had a heavy rain that night, but luckily it held up and we didn’t plug any drains. We had somebody there watching just in case all of our paper snow came down and plugged one of the drains,” Benjamin remarks.

Airbag decks were favored over hydraulic gimbals, such as when the school bus is tipping over the severely damaged Golden Gate Bridge. “We did one shaky gimbal set and that was for the hallway scene in the USSR when the ship was shaking,” Benjamin explains. “The monster comes in there and starts bashing around. We built that whole big deck on a floor and made it flip back and forth and tilted it up to whatever angle we wanted. We had big shaker motors on it and rocked it back and forth this way and tilted it that way. We shook that one pretty good.” It was not all about shaking things. “For the last few episodes we did a vortex, so we had a lot of big wind machines and ratcheting things and pulling things into the vortex. I want to see that portion of it. For the vortex in Alaska, we were shaking the trucks and equipment. But when the creature was chasing them, we didn’t do any explosions for that. The only thing that we did was blast air cannons to have some snow flying around. We did some pyro for the seismic charges that were set off at the power plant. Then we had Kurt Russell running through the lightning field. We had a bunch of mortars going off around them at that point, too.” After being in special effects for 23 years, Benjamin learned a particular lesson. “A lot of times, I find that the smaller gags are trickier than the bigger ones, like the dripping goo coming down from the ship. Something like that can be a lot of work, and testing to the desired look that everyone wants for that is sometimes more work than flipping a car over.” The practical elements are critical in making the stunts and visual effects believable. “We’re trying to do anything to help give the set some life so that the actors can get into it a bit more. You don’t have to fully act when you have the set moving around.”

Collaboration is pivotal to the success of any project. “Ultimately, my attitude is that visual effects are a big part of the show,” states Jess Hall, Cinematographer, Episodes 101 and 102. “The CG has to be integrated into the photography, so I take it as my responsibility that those things have to work together. That means being collaborative and also organized about how you light; for example, on greenscreen matching lighting and doing the work in advance in terms previs and storyboards. But, ultimately, I treat it as a collaboration for which I bare a lot of responsibility for the end result. It’s not like I’m going to shoot someone on a greenscreen, hand it over to visual effects and let them do their thing; that’s not going to produce a good result.” A different color palette was adopted for the series. “We scaled back a little bit on the gaudier and pulp elements of some of the movies. We tried to bring it more into the dramatic cinematic space. Even if you look at our version of Skull Island. my reference for that was more Apocalypse Now. It was the golden warm light but naturalistic approach. Then we go to Tokyo and you have these cool tones, but the lighting was always naturalistic. The composition was reasonably consistent. The lenses I would shoot the faces on were a similar composition. You’re building this thread of visual language that is bulletproof in a way. You can apply these period looks or more action-sequence elements in there, but it doesn’t feel out of place. That was the challenge of the show, and a lot of thinking around design for me was about bringing them together enough, but having them different enough because you also had to understand these timelines. It was important that Skull Island did look different from the 1950s content, otherwise I don’t think you understand where you were.”

Hall partnered with filmmaker Matt Shakman on Episodes 101 and 102. “Matt is a dramatist and a real actor’s director,” Hall describes. “Having to try to pretend that things are moving around you, and you’re on a bluescreen and saying, ‘Okay, the crowd is coming from the right.’ Or, ‘Now turn to look at the crowd 300 feet away.’ ‘Follow the tennis ball.’ We’ve all seen how that can be quite tough for actors and performers. Matt is always looking to put the actors in the position where they’re comfortable and can give the best performance, but also feel the scene. We went to practical photography. There were a lot of visual effects but a lot of photography on location, and a lot of in-camera stunt action and real effects work that went on in all of these scenes that added to the sense of realism, which is what we wanted.” Atmospherics like smoke were never utilized without intention. “You have to be careful about how much you put in because quickly you lose the contrast of the shot. It’s something that you rely on special effects to operate. Ultimately, I’m the one who has to say, ‘Turn the smoke on or off.’ The whole show had this low-level haze. I was going for a softer, more dramatic look with a bit of texture in there and in the shadows.” The legacy of what has come before loomed large over the production. Hall observes. “You’ve got this huge IP and franchise and so many different elements to it. How do you take that, respect that and do that justice, but also do something that is distinct and appropriate for the show that you’re making and is your own work. Threading that needle was hard.”